Medical Process Step Control With Individual Data-Stream Classifiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex processes, particularly in medical environments, are difficult to monitor and control due to variability in data streams provided by different systems, institutions, countries, and operation teams, making it challenging to ensure the right working steps are performed at the right time.
Innovation Solution
A method that utilizes individual classifiers trained on specific data streams to determine the probability of current and next working steps, using a combination of neural networks and decision trees to generate a control signal based on a list of possible working steps, adaptable to different institutions and countries, and incorporating an overview data stream to optimize process section determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a single unified system is used to monitor and control the entire process, then comprehensive process control is achieved, but the system complexity and difficulty of monitoring increase significantly
Solution Approach 1:
The patent divides the complex process monitoring system into multiple independent subsystems, each responsible for specific process sections. Each subsystem independently monitors and controls its assigned process section, preventing the need for a single complex centralized system while maintaining comprehensive process control.
Solution Approach 2:
The patent introduces a hierarchical dimension to the control architecture, with multiple independent subsystems operating at different process sections. This dimensional approach allows comprehensive monitoring without requiring a single complex system, as each subsystem operates autonomously within its domain.
2Adaptability or versatility
If multiple data streams from different systems and institutions are integrated, then process coverage is improved, but the difficulty of detecting and measuring process state increases
Solution Approach 1:
The patent segments the data processing task by assigning different data streams to different process sections. Each subsystem processes only the data streams relevant to its process section, reducing the complexity of detecting and measuring process state while maintaining comprehensive coverage across all sections.
Solution Approach 2:
The patent applies local quality by tailoring the data processing and detection methods to each specific process section. Each subsystem uses detection methods optimized for its local process characteristics, making process state detection easier while accommodating diverse data streams from different sources.
3Measurement precision
If individual classifiers are trained on specific data streams for each process section, then measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent segments the classifier system into multiple independent classifiers, each trained on specific data streams for particular process sections. This segmentation improves measurement precision for each section while managing overall system complexity through modular, independent classifier units.
Solution Approach 2:
The patent applies partial action by training classifiers only on the data streams and process sections they need to handle, rather than using a single classifier for all data. This selective approach improves precision for specific tasks while keeping the overall system complexity manageable through focused, specialized classifiers.
Data Source
AI summary
A computer-implemented method comprises: provisioning a plurality of data streams, each of the plurality of data streams being assigned an individual classifier; provisioning a list including a plurality of possible working steps; applying the plurality of individual classifiers to the plurality of data streams, wherein for each working step, based on the assigned data stream, a probability is determined; determining a current or next working step as a function of the probabilities; and provisioning the control signal.